aws-samples/sagemaker-cv-preprocessing-training-performance
SageMaker training implementation for computer vision to offload JPEG decoding and augmentations on GPUs using NVIDIA DALI — allowing you to compare and reduce training time by addressing CPU bottlenecks caused by increasing data pre-processing load. Performance bottlenecks identified with SageMaker Debugger.
This project helps machine learning engineers and data scientists accelerate computer vision model training on Amazon SageMaker. It takes your existing image datasets and training scripts and outputs faster training times and insights into CPU bottlenecks. This is especially useful for those working with large image datasets and complex augmentation pipelines.
No commits in the last 6 months.
Use this if you are a machine learning engineer experiencing slow training times for your computer vision models on Amazon SageMaker due to heavy image preprocessing and augmentations.
Not ideal if your training bottlenecks are not related to CPU-bound image decoding and augmentation, or if you are not using Amazon SageMaker for your model training.
Stars
21
Forks
2
Language
Python
License
—
Category
Last pushed
Jul 13, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/aws-samples/sagemaker-cv-preprocessing-training-performance"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
microsoft/onnxruntime
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
onnx/onnx
Open standard for machine learning interoperability
PINTO0309/onnx2tf
Self-Created Tools to convert ONNX files (NCHW) to TensorFlow/TFLite/Keras format (NHWC). The...
NVIDIA/TensorRT
NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This...
onnx/onnxmltools
ONNXMLTools enables conversion of models to ONNX